phase change references

Numerical methods and mathematical models of Elmer
Post Reply
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

phase change references

Post by NJank »

Wondering if anyone with a good archive might have academic references to the models used for the "Spatial 1", "Spatial 2", and "Temporal" forms of effective heat capacity approximation given in the models manual. Or maybe a good memory of which papers might have described the model choices used? I've been getting some very different responses to all three based on the things like time step, type of smoothing/approximation function used, etc. Was hoping to dive into some of the theory behind them, as I've only seen the simpler Spatial 1 (dH/dT) described in academic articles.
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

Re: phase change references

Post by NJank »

I apparently need to reread some of my old posts:

viewtopic.php?p=7624#p7624
"Spatial 2, sqrt((delH.delH)/(delT.delT)), I've seen referred to as the Lemmon scheme [1]... Lemmon, E.C., “Multidimensional integral phase change approximations for finite element conduction codes”, in Lewis, R.W., Morgan, K. and Zienkiewicz, O.C. (Eds), Numerical Methods in Heat Transfer, Wiley, Chichester, 1981, pp. 201-13."
looking for a more accessible reference, seems the U.S. Department of Energy has a decent archive server -
Lemmon, E. C. Phase-change techniques for finite element conduction codes. No. CONF-790710-2. Idaho National Engineering Lab., Idaho Falls (USA), 1979. https://www.osti.gov/servlets/purl/6202489

no real need for a reference for Spatial 1, since dH/dT is the default method. will see if it goes into the Temporal version at all, although that just seems to be a temporal equivalent of the Spatial 2 gradient scheme.
NJank
Posts: 99
Joined: 05 Dec 2009, 00:05
Location: Baltimore, MD, USA

Re: phase change references

Post by NJank »

a few more have come up. the following provides a summary of the three methods with some references others can follow:

Pham, Q. T. "Comparison of general-purpose finite-element methods for the Stefan problem." Numerical Heat Transfer 27.4 (1995): 417-435.
http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf

It describes Lemmon's spatially averaged method (Spatial 2) in equation 9, as well as the Time averaged method (Temporal), with some mention of the 'default' Spatial 1 method (dH/dT), in particular mentioning that it only really works well for smoothed effective capacity approximations.

they offer some guidelines on which to use when. One interesting statement:
Space-averaged specific heat methods are associated with distributed capacitances, while time-averaged specific heat methods must be used with lumped capacitances.
Post Reply